Developing a screening program for fabry disease using unexplained increased left ventricular wall thickness measured by echocardiography at an academic centre

Abstract Background Most cases of Left Ventricular Hypertrophy (LVH) are the result of well described disorders with known etiology but occasionally cases of unexplained LVH are discovered and the prognosis for these patients remains poor. Fabry Disease (FD) is in the differential diagnosis for LVH but unfortunately, its detection is difficult. Although these cases are rare it is reasonable to encounter these patients on an annual basis. Methods We proposed an algorithm (Figure 1) to screen all echocardiograms presenting to our centre with a Left Ventricular Posterior Wall Thickness (LVPWT) ≥ 14 mm or ≥ 15 mm in the presence of arterial hypertension, obesity, diabetes, and/or CKD stage ≥ 3. Those with known infiltrative cardiomyopathies and at least moderate aortic stenosis or mixed aortic valve disease were excluded. Results Approximately 10000 patients were referred for echocardiography in 2021 to our centre. 570 patients were identified with a LVPWT ≥ 14 mm. The echocardiograms were reviewed for quality control by two sonographers and two echocardiographers. 100 (17.4%) patients did not have images that were accessible for review. 180 (31.5%) patients were excluded due to revising the LVPWT to < 14 mm. 83 (14.5%) patients had passed away during the year. 82 patients (14.4%) had at least moderate aortic stenosis and/or mixed aortic valve disease. 20 (3.5%) patients carried a diagnosis of hypertrophic cardiomyopathy. 2 (0.35%) patients carried a diagnosis of FD. 7 (1.2%) patients had amyloidosis. 1 (0.17%) patient had Left Ventricular Non-Compaction. 1 patient had sarcoidosis (0.17%). 59 (10%) patients had arterial hypertension, obesity, diabetes, and/or CKD stage ≥ 3 or other life limiting processes with LVPWT < 15 mm. 27 patients had LVPWT ≥ 15 mm with arterial hypertension, obesity, diabetes, and/or CKD stage ≥ 3. 9 Patients had LVPWT ≥ 14 mm and no major comorbidities. Initially 36 (6.3%) patients were identified has having unexplained LVPWT. 7 were >80 years of age and 20 were noted to have a normal LVPWT upon follow up. In 21 image quality was sufficient for strain measurements to be performed in the LVPW of which 12 were noted to have a reduced strain. In total only 9 individuals (1.6%) were eligible for genetic screening. Conclusion Our algorithm shows that the accurate assessment of the LVPWT remains challenging and that this measure alone is not sufficient to appropriately screen for Fabry's disease. Our study suggests the need for caution the need in future AI algorithms driven by standard measurements alone. The incorporation of ECG & clinical findings may be more successful in identifying patients suitable for genetic testing than echocardiography measurements alone.